‘THIRD PERSON INDICATIVE /-s/: CONSTRAINTS ON SECOND LANGUAGE ACQUISITION P. Hironymous sere third person indicative [sis generally viewed as abasic intro id person indicative /-s/ is ly viewed as a basic introductory concept for second language (L2) learners of English, It can be found atthe beginning of almost any first year grammar text (Molinsky & Bliss, 1983; Azar, 1985; Elbaum, 1986) along with a straightforward explanation of its use. ‘The concept of /-s/ is both understandable and teachable, and the first few days of beginning level instruction are often filled with a variety of exercises, games, and communicative activities for first year students. Stephen Krashen (1982:17) labels the third person /-s/ as one of the "easy" forms, and most teachers would probably agree; it is considered so “easy” that, after those first few days, /-s is seldom the focus of any lesson again. It is clear to teachers of English as a Second Language (ESL), however, that few students actually seem to apply this simple concept consistently. Elaine Tarone agrees, noting that omission of this morpheme “continues to occur even in the production of advanced leamers" (1988:97). Problems with /-s/ might be expected in natural speech and free , but teachers are often perplexed when students fail to use /- s/ even after careful editing. Even more perplexing are students who will omit /-s/ while reading aloud; the /-s! here seemingly requires nothing more than pronunciation of the print letter. Researchers, ESL teachers, and students seem to agree that the rule of third person indicative /-s/ is simple, and the form is readily available in the language environment, ‘What then is causing varability in the use of third person indicative /-s7?T suggest that acquisition of third person /-s/ is not as "easy" as most believe. I propose that semantic and syntactic complexity in the context of general /-s/ and constraints in phonology, perhaps those dealing with universals, can inhibit complete acquisition for many L2 learners of English. Meaning and Syntax: Complexties of /-s/ ‘Brown's (1973) study of Adam, Eve, and Sarah acquiring English as a first language was one of the first to provide a rank order of morphemes acquired in children, Dulay and Burt (1980) followed by investigating eight of these ordered morphemes for Spanish speaking children and found a similar order. Of the eight morphemes investigated by Dulay and Burt, six were concemed with the use of /-s/: contractible copula (’s), plural (-8), contractible auxiliary (’s), possessive (’s) , third person (-s), and long plural (-s). Although Dulay and Bux were concered wit the sequision order ofthese morphemes and not semantics or syntax, their list points out the frequency and varieties of /-s/ encountered by 2 children in the initial stages of language acquisition. For L2 adults who are confronted with complex language input every day, of sorting out the above uses of /-s/ and other English morphemes must seem enormous, In examining short plural /-s/, for example, we can see that the L2 learners must eam certain syntactic and semantic constraints; short plural /-s/ must attach to the end of a noun, and /-s/ denotes plurality of that noun. Examining possessive /°s/ next to this, we see that it also attaches to the end of a noun, but it can denote plural or singular possession of the following noun. When the contractible copula /’s/ is acquired, the learner has to store this morpheme with similar formation rules to the above two, but the /’s/ here represents a reduction of a whole word, the copula is, and may be followed by adjectives, nouns, or prepositional phrases which describe the noun that the /’s/ has been attached to’. Sometime during or after the acquisition of the above morphemes, the third person indicative /-s/ is acquired (see discussion below). This use of /-s/ conflicts with the above uses in some important ways; it attaches only to verbs, and it denotes only singular to match a 2third person singular subject From the above examples, we can begin to see that the third person indicative /-a/ is hardly that when pt in in the content of just a few of the other conflicting uses of /-! in English To adult 12 earners of English, /-s/ seems to attach to nouns or verbs, denote plural, plural or singular, or singular, and be contextually related to constituents of either the noun or the verb phrase. Even ifthe /-s/ morphemes are not sequenced, but clustered in hierarchical groups as, Dulay and Burt propose, the L2 leamers are still faced with conflicts between and within acquired clusters (1980:356). Learners of English will eventually sort out the various uses for general /-s/ and even continue to add others not covered in Dulay and Bunt’s study, that is, the contractible auxiliary /’s / of third person auxiliary has + /-en/ and the contractible auxiliary /’s/ of third person indicative passive constructions. First language learners eventually acquire and use /-s/ without difficulty, but as noted earlier (Tarone, 1988:97), even very advanced L2 learners will still omit them at times. and syntactic complexity may account for some omissions of the various uses of /-s/, but it cannot account for variability within the use of one of these, namely the third indicative. Perhaps some aspects of the phonology of /-s/ may be able to account for this kind of variability. Phonology: Constraints on /-s/ ‘The continued variability of third person indicative /-s/ in 12 production of English may be related to phonological constraints concerned with the reduction of word-final consonant clusters and the influence of the sonority hierarchy. Fred Eckman collected and examined data conceming production of final consonant clusters of English spoken by native speakers of an L1 which does not allow these clusters (1987:144), In his study, he draws on universal generalizations (1987:151-52; from Greenberg, 1978) as the basis for proposing the following optional cluster reduction rule in (1)(=Eckman’s [7}). Eckman admits that the rule as stated does not predict which consonant is deleted and adds a markedness constraint, in (2). (1) Cluster Reduction (CR) (optional) CCOF-CO# Optionally delete one member of a first tr-literal consonant cluster; optionally Galete one member of a final brliteral consonant cluster, (1987.12) Q) Marked Cluster Constraint (MCC) Given the set CC, which contains all logically-possible bi-literal clusters obtainable by applying CR to a final cluster, a cluster CC, which is a member of CC, will be an output of CR only if CC, is not more ‘marked relative to the other members of CC... (1987:154) He found, however, that several forms produced by the L2 leamers of English violated the MCC; among the forms were final clusters of fricative reduced to stop-stop, as in the following examples of third person indicative /-s/, in (3). @) accepts _—_Jaksepts/ -» /aksept/ acts Teekts/ -> /ekt/ Although Eckman is concerned with a much larger issue than this in his study, his comments are pertinent since they focus on /-s/ as a morpheme. When L2 speakers of English were given a list of words to pronounce, Eckman reports that the fina] cluster was reduced to stop-stop, and it “almost invariably involve{d] the deletion of a final fricative; that is either the plural or third person marker." (See Appendix A.) He further states that“there are no instances in which a stop-stop cluster is produced by deleting /s/ where it is not a morpheme" (1987:154). Native speakers, however, do not reduce consonant clusters in this way; Eckman reports, for example, that the L1 English speakers may pedace opts i retained and Japts/ to [aps] where the stop is deleted but the fricative would morpheme kept inact This brings us to the interesting conclusion that there is ing about the very nature of /-s/ as a morpheme which will act to preserve /-s/ in the first but delete it in L2 English. Only when /-s/ is a morpheme, such as the third person ive /-s/ in Eckman’s study, will L2 leamers delete it and thus violate Eckman’s proposed rules for eduction and markedness. Eckman is careful to describe his CR rule as optional and his markedness constraints as only tendencies in interlanguage (IL), but if they do as he claims in this study, then, we see that the deletion of /-s/ by L3 learners of English violates ‘Ecknan’s CR rule and markedness constraints. According to Eckman, the L2 speakers should preserve it. Why are they so eager to delete it? The answer does not seem to lie within Eckman’s study of consonant reduction and markedness constraints, but perhaps Phonology of third person indicative /-s/ with regard to the sonority hierarchy might provide more insights into his data. AS seen above, L2 leamers of English will allow final consonant clusters even though their languages may not, but they seem to violate Eckman’s proposed universal constraints by deleting third person indicative /-s/._A study by Herbert Tropf (1987) might provide some reasons for these deletions. Tropf proposes that the sonority hierarchy may influence consonant clusters and syllable structure; his summary of the sonority hierarchy follows in 5) and is depicted in (6), (S) Each syllable has a peak, mostly a vowel. If additional segments are present at the beginning of the end of the syllable, then they have a Some orem namie eee es syllable peak to ipheral segments. , 1987:175; cf. Hooper, 1976; Kiparsky, 1979; Sele, 1984) © Sonority Hierarchy (Tropf, 1987:176): c c N L Gv G@kLN cic plosive friative nasal liquid glide vowal glide ‘Liquid naaal_—_fricaive.plosive (less sonorous) «~—-~--------- (most sonorous) ——-——--~ + (less sonorous) In a study of Spanish L1 ledrning German as an L2, Tropf investigated whether sonority affects the L2 phonological acquisition and if this can explain "certain “free” variants which cannot be explained through [native language] or {target language] restrictions” (1987:177). Since Tropf’s study does not investigate English as an L2, third person indicative /-s/ is not addressed directly. However, from Tropf’s demonstration that the sonority hierarchy contributes to vz ity in consonant clusters, we can infer similar effects on consonant clusters in English. With to omission of syllable final consonants, Tropf reports that "the more target consonant, the more Likely iti realized a a (possibly deviant) the IL, but not omitted ... a similar observation holds for final target clusters” (1987:188). Given this information, we can return to the data from Eckman’s study and examine examples from those data which contain final clusters formed from adding third person indicative /-s/. Eckman’s list contains almost thirty third person indicative verbs, six of which are listed in (7)(taken from Appendix A), representing each of the final consonant clusters involved. In each case, the sonority hierarchy is violated in the final consonant 4cluster by addition of the fricative morpheme /-s/. @M claps c G vocoec op glide vowel map —fiouive eats veocioc vowsl sop fietve cooks co vecoc vowel sap —“fintve asks War Clce C vowel fictive sop fiative tests c voc cc op vowel fictive nop fictive acts vococeec vowel op top fictive sifts c vec cc ftioaive vowel fictive sop fictive accepts = C cc ¢ fative vend mp top fionive Considering Eckman’s data in light of Tropf’s study, it becomes clear that there is a tendency for these L2 leamers of English to omit /-s/ when it violates the sonority hierarchy, thereby overriding Eckman’s consonant reduction and markedness constraints. If the sonority hierarchy is a universal, then it would seem to be a powerful constraint which can act to inhibit complete acquisition of third person indicative /-s/ for many L2 leamers of English, ‘This may answer the orginal question conceming the variability of the third son indicative /-s/ by even advanced L2 leamers; they may delete third person indicative -s/ when it violates the sonority hierarchy, but retain it when it does not. Conclusion: A proposed study of /-s/ The above claim is obviously open to empirical investigation. After L2 leamers of English have sorted out the semantic and syntactic complexities of general /-s/ in English and begin to use third person indicative /-s/, we can investigate the variability of this morpheme in a variety of language functions. Through free writing and editing, reading aloud, and natural speech we can examine learners’ omissions of /-8/ from final consonant clusters in relation to the sonority hierarchy. If we find that /-s/ is retained in vowel final verbs, for example, but deleted after stops, we may be able to explain the variability in terms of sonority constraints. As Tropf pointed out in his investigation of an interlanguage, variations which are not explained by the function or by the native or target language may show that "a universal factor is operative" (1987:174-175). If we can demonstrate that the sonrity hierarchy opersies to inhibit productions of third person indicative /-s, then this “easy” rule in English could be under a powerful universal constraint; the inherent tendency of /-s/ to violate a possible universal such as the sonority hierarchy can make third perso indicative /-s/ one of the most difficult things for many adult L? learners of English to acquire. Azar, B. 1985. Fundamentals ge tk grammar. Englewood Cliffs, NJ: Prentice-Hall. Brown, R. 1973. A first lans bridge, MA: Harvard Univ. Press. Bar, Mand Dulay, H. 1980. On acquisition orders. Ins. Felix (Ed. Second language Vievelopment trends and issues. Tubingen: Gunter Narr Verlag. Eckman, R, 1987. The reduction of word-final consonant clusters in interlanguage. In A.James and J. Leather (Eds.), Sound patterns in second language acquisition. Dordrecht: Foris. Elbaum, S. 1986. Grammar in context--Book 1. New York: HarperCollins. Greenberg, J. 1978. Some generalizations concerning initial and final consonant clusters. In J. Greenberg, C. Ferguson and E. Moravesik (Eds.), Universals of human language. Stanford, CA: Stanford Univ. Press. Hooper, J. 1976. An introduction to natural generative phonology. New York: Academic Press. Kiparsky, P. 1979. Metrical structure assignment is cyclic. Linguistic Inquiry 10:421-441. Krashen, 8. 1982, Principles and practice in second language acquistion. Oxford: Pergamon Press. Molinsky, S. & Bliss, B. 1983. Line by line: English through GrammarStories. Englewood liffs, NJ: Prentice-Hall. Selkirk, E. 1984. Phonology and syntax: the relation berween sound and structure. Cambridge, MA: MIT Press. Tarone, E. 1988. Variation in interlanguage. London: Edward Amold. Tropf, H. 1987. Sonority as a variability factor in second language phonology. In A. James and J. Leather (Eds.), Sound patterns in second language acquisition. Dordrecht: Foris.APPENDIX A. WORD LIST Final Clusters (Eckman, 1987:160) st sts test erased costs nest released tests post passed nests dust missed posts cost bussed sits sk sks ks ask husks box bask asks tax husk basks fix mask ‘tusks wax tusk masks mix skt it masked duct tucked asked fact packed tusked pact backed basked act booked hhusked sect cooked chsp gasps, ped grasps rasp clasps gasped gasp rasps rasped lisp fees grasped grasp isps. clasped pt Tips ‘opt claps adopt s sept $ accept gaps adept BES eu 55585